- 15 Aug, 2017 1 commit
-
-
Michael Niedermayer authored
Reviewed-by: atomnuker Signed-off-by:
Michael Niedermayer <michael@niedermayer.cc>
-
- 21 Mar, 2017 2 commits
-
-
James Almer authored
Fixes memleak. Reviewed-by:
wm4 <nfxjfg@googlemail.com> Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
James Almer <jamrial@gmail.com>
-
James Almer authored
Also make it more readable while at it. Signed-off-by:
James Almer <jamrial@gmail.com>
-
- 17 Mar, 2017 1 commit
-
-
Muhammad Faiz authored
when set_compensation is called with zero sample_delta, compensation does not happen (because dst_incr == ideal_dst_incr) but compensation_distance is set regression since 01ebb57cFound-by:
wm4 <nfxjfg@googlemail.com> Reviewed-by:
wm4 <nfxjfg@googlemail.com> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 09 Mar, 2017 1 commit
-
-
Muhammad Faiz authored
except filter_length == 1 odd filter_length gives worse frequency response, even when compared with shorter filter_length also makes build_filter simpler Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 08 Mar, 2017 1 commit
-
-
Muhammad Faiz authored
this gives better frequency response update swresample fate and other fates that depend on resampling Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 13 Dec, 2016 2 commits
-
-
Muhammad Faiz authored
integrate it inside multiple_resample allow some calculations to be performed outside loop Suggested-by:
Michael Niedermayer <michael@niedermayer.cc> Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
Muhammad Faiz authored
This should fix Ticket6012 Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 03 Dec, 2016 1 commit
-
-
Muhammad Faiz authored
Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 24 Nov, 2016 1 commit
-
-
Muhammad Faiz authored
separate dsp.resample to dsp.resample_common and dsp.resample_linear and choose to call faster resample_common even when linear_interp=on when c->frac and c->dst_incr_mod are both zero speed up resampling when exact_rational and linear_interp are both enabled because exact_rational force c->frac and c->dst_incr_mod to be zero when soft compensation does not happen benchmark on exact_rational=on:linear_interp=on old new real 8.432s 5.097s user 7.679s 4.989s sys 0.125s 0.107s Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 18 Oct, 2016 1 commit
-
-
Muhammad Faiz authored
return AVERROR code on error Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 20 Jun, 2016 1 commit
-
-
Muhammad Faiz authored
phase_shift and phase_mask is removed generally exact_rational=on is faster than exact_rational=off Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 17 Jun, 2016 2 commits
-
-
Muhammad Faiz authored
high phase_count is only useful when dst_incr_mod is non zero in other word, it is only useful on soft compensation on init, it will build filter with low phase_count but when soft compensation is enabled, rebuild filter with high phase_count this approach saves lots of memory Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
Muhammad Faiz authored
because exact_rational does not guarantee that phase_count is even Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 13 Jun, 2016 1 commit
-
-
Muhammad Faiz authored
give high quality resampling as good as with linear_interp=on as fast as without linear_interp=on tested visually with ffplay ffplay -f lavfi "aevalsrc='sin(10000*t*t)', aresample=osr=48000, showcqt=gamma=5" ffplay -f lavfi "aevalsrc='sin(10000*t*t)', aresample=osr=48000:linear_interp=on, showcqt=gamma=5" ffplay -f lavfi "aevalsrc='sin(10000*t*t)', aresample=osr=48000:exact_rational=on, showcqt=gamma=5" slightly speed improvement for fair comparison with -cpuflags 0 audio.wav is ~ 1 hour 44100 stereo 16bit wav file ffmpeg -i audio.wav -af aresample=osr=48000 -f null - old new real 13.498s 13.121s user 13.364s 12.987s sys 0.131s 0.129s linear_interp=on old new real 23.035s 23.050s user 22.907s 22.917s sys 0.119s 0.125s exact_rational=on real 12.418s user 12.298s sys 0.114s possibility to decrease memory usage if soft compensation is ignored Signed-off-by:
Muhammad Faiz <mfcc64@gmail.com>
-
- 16 May, 2016 1 commit
-
-
Michael Niedermayer authored
Signed-off-by:
Michael Niedermayer <michael@niedermayer.cc>
-
- 24 Dec, 2015 1 commit
-
-
Ganesh Ajjanagadde authored
Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
- 15 Nov, 2015 1 commit
-
-
Ganesh Ajjanagadde authored
It is inherently double precision, and 1.0 is perfectly represented anyway. Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
- 11 Nov, 2015 1 commit
-
-
Michael Niedermayer authored
Signed-off-by:
Michael Niedermayer <michael@niedermayer.cc>
-
- 09 Nov, 2015 4 commits
-
-
Ganesh Ajjanagadde authored
This may be a slightly surprising optimization, but is actually based on an understanding of how math libraries compute trigonometric functions. Explanation is given here so that future development uses libm more effectively across the codebase. All libm's essentially compute transcendental functions via some kind of polynomial approximation, be it Taylor-Maclaurin or Chebyshev. Correction terms are added via polynomial correction factors when needed to squeeze out the last bits of accuracy. Lookup tables are also inserted strategically. In the case of trigonometric functions, periodicity is exploited via first doing a range reduction to an interval around zero, and then using some polynomial approximation. This range reduction is the most natural way of doing things - else one would need polynomials for ranges in different periods which makes no sense whatsoever. To avoid the need for the range reduction, it is helpful to feed in arguments as close to the origin as possible for the trigonometric functions. In fact, this also makes sense from an accuracy point of view: IEEE floating point has far more resolution for small numbers than big ones. This patch does this for the Blackman-Nuttall filter, and yields a non-negligible speedup. Sample benchmark (x86-64, Haswell, GNU/Linux) test: fate-swr-resample-dblp-2626-44100 old: 18893514 decicycles in build_filter (loop 1000), 256 runs, 0 skips 18599863 decicycles in build_filter (loop 1000), 512 runs, 0 skips 18445574 decicycles in build_filter (loop 1000), 1000 runs, 24 skips new: 16290697 decicycles in build_filter (loop 1000), 256 runs, 0 skips 16267172 decicycles in build_filter (loop 1000), 512 runs, 0 skips 16251105 decicycles in build_filter (loop 1000), 1000 runs, 24 skips Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
Ganesh Ajjanagadde authored
When upsampling, factor is set to 1 and sines need to be evaluated only once for each phase, and the complexity should not depend on the number of filter taps. This does the desired precomputation, yielding significant speedups. Hard guarantees on the gain are not possible, but gains themselves are obvious and are illustrated below. Sample benchmark (x86-64, Haswell, GNU/Linux) test: fate-swr-resample-dblp-2626-44100 old: 29161085 decicycles in build_filter (loop 1000), 256 runs, 0 skips 28821467 decicycles in build_filter (loop 1000), 512 runs, 0 skips 28668201 decicycles in build_filter (loop 1000), 1000 runs, 24 skips new: 14351936 decicycles in build_filter (loop 1000), 256 runs, 0 skips 14306652 decicycles in build_filter (loop 1000), 512 runs, 0 skips 14299923 decicycles in build_filter (loop 1000), 1000 runs, 24 skips Note that this does not statically allocate the sin lookup table. This may be done for the default 1024 phases, yielding a 512*8 = 4kB array which should be small enough. This should yield a small improvement. Nevertheless, this is separate from this patch, is more ambiguous due to the binary increase, and requires a lut to be generated offline. Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
Ganesh Ajjanagadde authored
This improves accuracy for the bessel function at large arguments, and this in turn should improve the quality of the Kaiser window. It also improves the performance of the bessel function and hence build_filter by ~ 20%. Details are given below. Algorithm: taken from the Boost project, who have done a detailed investigation of the accuracy of their method, as compared with e.g the GNU Scientific Library (GSL): http://www.boost.org/doc/libs/1_52_0/libs/math/doc/sf_and_dist/html/math_toolkit/special/bessel/mbessel.html. Boost source code (also cited and licensed in the code): https://searchcode.com/codesearch/view/14918379/. Accuracy: sample values may be obtained as follows. i0 denotes the old bessel code, i0_boost the approach here, and i0_real an arbitrary precision result (truncated) from Wolfram Alpha: type "bessel i0(6.0)" to reproduce. These are evaluation points that occur for the default kaiser_beta = 9. Some illustrations: bessel(8.0) i0 (8.000000) = 427.564115721804739678191254 i0_boost(8.000000) = 427.564115721804796521610115 i0_real (8.000000) = 427.564115721804785177396791 bessel(6.0) i0 (6.000000) = 67.234406976477956163762428 i0_boost(6.000000) = 67.234406976477970374617144 i0_real (6.000000) = 67.234406976477975326188025 Reason for accuracy: Main accuracy benefits come at larger bessel arguments, where the Taylor-Maclaurin method is not that good: 23+ iterations (at large arguments, since the series is about 0) can cause significant floating point error accumulation. Benchmarks: Obtained on x86-64, Haswell, GNU/Linux via a loop calling build_filter 1000 times: test: fate-swr-resample-dblp-44100-2626 new: 995894468 decicycles in build_filter(loop 1000), 256 runs, 0 skips 1029719302 decicycles in build_filter(loop 1000), 512 runs, 0 skips 984101131 decicycles in build_filter(loop 1000), 1024 runs, 0 skips old: 1250020763 decicycles in build_filter(loop 1000), 256 runs, 0 skips 1246353282 decicycles in build_filter(loop 1000), 512 runs, 0 skips 1220017565 decicycles in build_filter(loop 1000), 1024 runs, 0 skips A further ~ 5% may be squeezed by enabling -ftree-vectorize. However, this is a separate issue from this patch. Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
Ganesh Ajjanagadde authored
Kaiser windows inherently don't require beta to be an integer. This was an arbitrary restriction. Moreover, soxr does not require it, and in fact often estimates beta to a non-integral value. Thus, this patch allows greater flexibility for swresample clients. Micro version is updated. Reviewed-by:
Derek Buitenhuis <derek.buitenhuis@gmail.com> Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
- 06 Nov, 2015 1 commit
-
-
Ganesh Ajjanagadde authored
This uses the trigonometric double and triple angle formulae to avoid repeated (expensive) evaluation of libc's cos(). Sample benchmark (x86-64, Haswell, GNU/Linux) test: fate-swr-resample-dblp-44100-2626 old: 1104466600 decicycles in build_filter(loop 1000), 256 runs, 0 skips 1096765286 decicycles in build_filter(loop 1000), 512 runs, 0 skips 1070479590 decicycles in build_filter(loop 1000), 1024 runs, 0 skips new: 588861423 decicycles in build_filter(loop 1000), 256 runs, 0 skips 591262754 decicycles in build_filter(loop 1000), 512 runs, 0 skips 577355145 decicycles in build_filter(loop 1000), 1024 runs, 0 skips This results in small differences with the old expression: difference (worst case on [0, 2*M_PI]), argmax 0.008: max diff (relative): 0.000000000000157289807188 blackman_old(0.008): 0.000363951585488813192382 blackman_new(0.008): 0.000363951585488755946507 These are judged to be insignificant for the performance gain. PSNR to reference file is unchanged up to second decimal point for instance. Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
- 04 Nov, 2015 1 commit
-
-
Ganesh Ajjanagadde authored
This speeds up build_filter by ~ 50%. This gain should be pretty consistent across all architectures and platforms. Essentially, this relies on a observation that the filters have some even/odd symmetry that may be exploited during the construction of the polyphase filter bank. In particular, phases (scaled to [0, 1]) in [0.5, 1] are easily derived from [0, 0.5] and expensive reevaluation of function points are unnecessary. This requires some rather annoying even/odd bookkeeping as can be seen from the patch. I vaguely recall from signal processing theory more general symmetries allowing even greater optimization of the construction. At a high level, "even functions" correspond to 2, and one can imagine variations. Nevertheless, for the sake of some generality and because of existing filters, this is all that is being exploited. Currently, this patch relies on phase_count being even or (trivially) 1, though this is not an inherent limitation to the approach. This assumption is safe as phase_count is 1 << phase_bits, and is hence a power of two. There is no way for user API to set it to a nontrivial odd number. This assumption has been placed as an assert in the code. To repeat, this assumes even symmetry of the filters, which is the most common way to get generalized linear phase anyway and is true of all currently supported filters. As a side note, accuracy should be identical or perhaps slightly better due to this "forcing" filter symmetries leading to a better phase characteristic. As before, I can't test this claim easily, though it may be of interest. Patch tested with FATE. Sample benchmark (x86-64, Haswell, GNU/Linux): test: swr-resample-dblp-44100-2626 new: 527376779 decicycles in build_filter(loop 1000), 256 runs, 0 skips 524361765 decicycles in build_filter(loop 1000), 512 runs, 0 skips 516552574 decicycles in build_filter(loop 1000), 1024 runs, 0 skips old: 974178658 decicycles in build_filter(loop 1000), 256 runs, 0 skips 972794408 decicycles in build_filter(loop 1000), 512 runs, 0 skips 954350046 decicycles in build_filter(loop 1000), 1024 runs, 0 skips Note that lower level optimizations are entirely possible, I focussed on getting the high level semantics correct. In any case, this should provide a good foundation. Reviewed-by:
Michael Niedermayer <michael@niedermayer.cc> Signed-off-by:
Ganesh Ajjanagadde <gajjanagadde@gmail.com>
-
- 07 Oct, 2015 2 commits
-
-
Michael Niedermayer authored
About 10% faster Signed-off-by:
Michael Niedermayer <michael@niedermayer.cc>
-
Michael Niedermayer authored
speedup of about 1% Signed-off-by:
Michael Niedermayer <michael@niedermayer.cc>
-
- 04 Jun, 2015 2 commits
-
-
Michael Niedermayer authored
Found-by:
wm4 <nfxjfg@googlemail.com> Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
Michael Niedermayer authored
Previous version reviewed-by: Pavel Koshevoy <pkoshevoy@gmail.com> Previous version reviewed-by: wm4 <nfxjfg@googlemail.com> Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 14 Mar, 2015 1 commit
-
-
Michael Niedermayer authored
Found-by: Clang -fsanitize=shift Reported-by:
Thierry Foucu <tfoucu@google.com> Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 16 Oct, 2014 1 commit
-
-
Michael Niedermayer authored
Fixes: asan_heap-uaf_2071250_7_139.ogg Fixes: assertion failure Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 03 Jul, 2014 1 commit
-
-
James Almer authored
Signed-off-by:
James Almer <jamrial@gmail.com> Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 02 Jul, 2014 1 commit
-
-
James Almer authored
The swresample_ prefix is not for internal functions Signed-off-by:
James Almer <jamrial@gmail.com> Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 28 Jun, 2014 1 commit
-
-
Ronald S. Bultje authored
Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 18 Jun, 2014 1 commit
-
-
Ronald S. Bultje authored
Also fix a bug with resample_compensation resetting dst_incr. Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 14 Jun, 2014 3 commits
-
-
Ronald S. Bultje authored
DSP bits of swri_resample go into their own mini-DSP functions; DSP init goes from a per-call branch in multiple_resample to a proper DSP init routine; x86 bits go into x86/; swri_resample() moves out of resample_template.c into resample.c because it's independent of DSP code or sample type; multiple_resample() is simplified. Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
Michael Niedermayer authored
Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
Ronald S. Bultje authored
Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 09 Jun, 2014 1 commit
-
-
Ronald S. Bultje authored
Signed-off-by:
Michael Niedermayer <michaelni@gmx.at>
-
- 02 Jun, 2014 1 commit
-
-
Ronald S. Bultje authored
I think there's an off-by-one in terms of the switchpoint where we switch from dst_incr to ideal_dst_incr, I don't think that's a massive issue, but just be aware of that. It's probably trivial to prevent but I don't care. Signed-off-by:
Michael Niedermayer <michaelni@gmx.at> I could not reproduce any off by 1 error, results are bit exact (michael)
-